Working and earning under different rules: what the United States can learn from labor market institutions in other developed countries.

Freeman, Richard B.

Working and Earning Under Different Rules: What the United States Can Learn from Labor Market Institutions in Other Developed Countries

Labor markets are the most idiosyncratic feature of developed countries, differing across national lines in more ways than markets for goods, finance, and the like. Some advanced capitalist countries are highly unionized; others are largely nonunion. Some countries rely on specific institutions to set pay: courts determine pay increases in Australia; the government extends collective bargaining agreements throughout the economy in Germany; in Japan, a large proportion of pay consists of bonuses, and the Shunto Offensive sets standards for wage increases throughout the economy. The United States, Canada, and the United Kingdom rely more on the free operation of markets.

There is a similar range of variation among countries in programs and institutions to aid those without work. Some countries have extensive income maintenance, social welfare, and labor market mobility and training programs, devoting considerable resources to redistributing income. Sweden is the typical example, but throughout Western Europe unemployment insurance systems have been more generous than the American system. The United States has relatively limited social welfare programs for most workers and for those without work, and relies to an unprecedented extent on the private charitable sector to deal with many social problems.

One of the most striking features of the 1980s was the substantial divergence in institutions and outcomes among developed western countries, often in ways that contravened the patterns of previous postwar history. The United States, which traditionally had higher rates of unemployment but relatively shorter durations of joblessness than Europe, experienced a "jobs explosion" with resultant lower unemployment than Western Europe. At the same time, the United States suffered rising inequality in earnings and poverty rates that eventually exceeded those in many other developed countries, and lost much of its lead in productivity and real earnings per person. The proportion of the work force organized in trade unions fell dramatically in the United States while remaining at higher levels in many other countries, including Canada. In earlier decades, the United States led the world in reducing working time toward the 40-hour week and extending vacation and holiday pay, but in the 1980s American workers put in more hours than did workers in Europe.

When unemployment rates rose in Europe in the early 1980s, most discussion of the difference between Western European and American labor markets focused on the advantages of labor arrangements of the U.S. style. The buzzword was "flexibility" of markets. Some analysts argued instead for "neocorporatist" arrangements, such as Sweden's, with central labor unions, employer confederations, and the state regulating the labor market. But most discussion focused on the problems of unemployment insurance systems that grant high benefits for years, long-term labor contracts that make firing workers difficult and hiring presumably more expensive, and so on. There was much concern with what Western Europe could learn from the flexible arrangements of the United States, but little thought as to what the United States might learn from European labor institutions.

The situation is the early 1990s looks quite different. There are rising inequality, high rates of poverty, particularly among children, homelessness, and related social problems in the United States. Unemployment rates, while still moderate, are now higher than those in West Germany. Widespread concern about the quality of the American labor force and American managerial practices has raised serious doubts about the way the United States has responded to the economic changes of the past decade or so. From an American perspective, the question is how foreign countries avoided some of our problems. We now need to ask what we can learn from our OECD peers to improve our society and our competitive economic performance.

The NBER's labor program has begun a large research endeavor to understand the operation of labor markets and income maintenance/safety net programs in other countries. We began with a research project comparing the United States with Canada. Ultimately there will be research projects on different aspects of labor relations and income maintenance programs in the United States and other OECD countries:

1) The Operation of Works Councils

Within Enterprises

2) Private Firm Training of Workers

3) Wage Structures and Alternate

Wage-Setting Systems

4) Income Maintenance and Social Welfare Programs

5) Programs to Aid the Extremely Poor

Some Early Results

A key area in which the U.S. labor market has diverged from many other OECD countries is the representation of labor at enterprises and in national economic decisionmaking. While the decline in union density is not unique to the United States - density dropped in the United Kingdom during Mrs. Thatcher's term in office, in the Netherlands, in Japan, and in France - nowhere has density fallen as much as in the private sector of the United States, and it has remained high in many other OECD countries, such as Germany, Belgium, Denmark, Sweden, and most notably Canada.

At the same time, most European countries have developed and strengthened the powers of "works councils," designed to allow management and labor to treat local workplace problems cooperatively. The United States, by contrast, has come to rely increasingly on the competitive market and on legal means of regulating local workplace arrangements, through legislation and judicial intervention. Consistent with the mixed results of the U.S. system, the European experience with quality-of-work circles and other innovative labor practices is that for labor-management cooperative arrangements to work, workers must be given some genuine authority or influence over decisions, by legislation or through union activity. The interesting question, on which research has yet to speak, is the extent to which these differing practices affect productivity and worker well-being in, say, similar plants of comparable multinational firms across the countries.

Although earnings differentials between more and less skilled or educated labor, and equality among workers, appear to be rising in most developed countries - presumably as a result of shifts in demand based on technological and other factors, and to the growth or world trade - the extent of the increase seems less extensive elsewhere. In Canada, wage differentials between more and less educated workers barely increased in the 1980s, while the differentials skyrocketed in the United States. One reason is that the number of college graduates grew more rapidly in Canada than in the United States in the 1980s. In Sweden, earnings differentials also rose. Swedish employers, unions, and workers came to realize that wage differentials had been compressed too much to make economic sense, and centralized collective bargaining broke down. However, the levels of differentials and increases are magnitudes different from those in the United States. Only in the United Kingdom have earnings differentials seemingly risen as much as in the United States.

Canada's experienced in the 1980s, when rates of poverty, in particular among children, fell below those in the United States, shows that alternative income maintenance programs in very similar economies can produce substantially different outcomes. The Canadian systems of child support, unemployment insurance, and health insurance appear to have worked better in "leaning against the wind" of rising inequality than comparable American programs. While virtually all research shows that there is a substantial labor supply response to providing benefits for those who are not working, extending the durations of unemployment benefits and linking benefits to work - "workfare" - appears to induce high labor participation, particularly of women. The Swedish welfare system is largely a workfare system. The relatively generous unemployment insurance entitlements in Canada appear to have induced a substantial number of women, who otherwise would not have worked, to work for at least part of the year.

Surprisingly, countries with very different labor market arrangements - the United States, with its decentralized, largely nonunion system, and Sweden, with its highly coordinated union system - had quite similar macroeconomic outcomes, with better employment experiences but productivity and real wage experiences that were relatively poorer than countries with less "extreme" arrangements. The implication is that there may be quite different paths to the same macroeconomic outcomes, even though the systems produce different microeconomic outcomes.

Overall, the NBER project on "working and earnings under different rules" will widen the range of economic experiences and data on which labor economists and others test theories and derive generalizations about the operation of market economies. The project should increase our understanding of which aspects of economic behavior are universal and which are influenced by laws, culture, and so forth. While the project is empirical at its core, it has a potentially close tie with modern game - theoretic analyses of markets that show that rational behavior can lead to very different outcomes depending on the precise rules of the game. The question of how different labor practices fit together into a "system" - for instance, whether one can adapt European works council or German apprenticeship programs, or Japanese style labor relations, to other countries - also calls out for theory as well as evidence.